spe113269 effective eor decision strategies with limited data manrique

8/11/2019 Spe113269 Effective EOR Decision Strategies With Limited Data Manrique

1/16

SPE 113269

Effective EOR Decision Strategies with Limited Data: Field CasesDemonstrationEduardo Manrique, SPE, Mehdi Izadi, SPE, Curtis Kitchen, SPE, Norwest-Questa Engineering and Vladimir

Alvarado, SPE, University of Wyoming

Copyright 2008, Society of Petroleum Engineers

This paper was prepared for presentation at the 2008 SPE /DOE Improved Oil Recovery Symposium held in Tulsa, Oklahoma, U.S.A., 1923 April 2008. This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not beenreviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, itsofficers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission toreproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

AbstractEnhanced-Oil Recovery (EOR) for asset acquisition or rejuvenation involves intertwined decisions. In this sense, EORoperations are tied to a perception of high investments that demand EOR workflows with screening procedures, simulationand detailed economic evaluations. Procedures have been developed over the years to execute EOR evaluation workflows.

We propose strategies for EOR evaluation workflows that account for different levels of available information. These procedures have been successfully applied to oil property evaluations and EOR applicability in a variety of oil reservoirs.The methodology relies on conventional and unconventional screening methods, emphasizing identification of analogues tosupport decision making. The screening phase is combined with analytical or simplified numerical simulations to estimatefull-field performance while maintaining rational reservoir segmentation procedures.

This paper fully describes the EOR decision-making procedures using field case examples from Asia, Canada, Mexico,South America and the United States. The type of assets evaluated includes a spectrum of reservoir types, from oil sands tolight oil reservoirs. Different stages of development and information availability are discussed. Results show the advantageof flexible decision-making frameworks that adapt to the volume and quality of information by formulating the correct

decision problem and concentrate on projects and/or properties with apparent economic merit.Our EOR decision-making approaches integrate several evaluation tools, publicly or commercially available, whosecombination depends on availability and quality of data. The decision is laid out using decision-analysis tools coupled witheconomic models and numerical simulation. This allows integrated teams to collaborate in the decision making processwithout over-analyzing the available data. One interesting aspect is the combination of geologic and engineering data,minimizing experts' bias and combining technical and financial figures of merit rationally. The proposed methodology has

proved useful to screen and evaluate projects/properties very rapidly, identifying whether or not upside potential exists.

IntroductionThis paper illustrates a set of strategies developed over a number of years to deal with improved/enhanced-oil recovery(IOR/EOR) decisions. In this paper, we are not concerned as to what IOR or EOR are or are not. Almost any strategy thatleads to increase recovery is under consideration for EOR/IOR decisions. The current oil market has triggered a significantincrease in property acquisition and the launching of enhanced-oil recovery projects. Tied to this competitive scenario is thescarcity of a properly trained workforce to deal with some decision challenges. In this scenario, the saying Time is moneycould not more factual. This will be referred to as one of the time constraints that decision makers face.

Excessive emphasis on prescriptive approaches to decision analysis in the absence of context bypasses the real purpose ofdecision analysis. What decision makers require is recommendations (support) based on realistic information on the state ofknowledge. Lack of context in decision making can be particularly effective in destroying value in development plans basedon EOR strategies. Inadequate focus can cause decision delays, which in turn lead to the risk of losing windows ofopportunity. In practice , reservoirs do not remain static during any exploitation phase, even if we do not do anything . Thelatter is one of the possible time constraints for a decision. Bidding for property acquisition is another. On the other hand, ithas been shown that falling into the over-analysis trap actually destroys value, because chances for success can diminish ifthe number of decisions is perilously limited (Begg and Bratvold, 2003).

One of the reasons for attachment to over-analysis probably comes from a deeply rooted belief in the Oil and Gas industrythat analysis through modeling can reduce uncertainty. The overconfidence bias associated with the belief that numericallyaccurate reservoir dynamic models can overcome the hurdles of ambiguity or even uncertain data sources is baseless. The


2/16

2 SPE 113269

model should only be as complex as necessary to answer the questions posed by the decision makers, i.e. the actual questionformulated by decision makers, assisted by support teams, in the opinion of the authors. Context matters for the formulationof decision strategies.

This paper illustrates pragmatic approaches to assisting decision-making exercises in real field situations. Emphasis is puton framing decisions and adaptation to information sources. However, nothing in this paper implies that complex modelingefforts are unnecessary, only that complexity should match the available data, time constraints and context of the decision. Anumber of screening strategies developed as a matter of necessity are proposed as key to make progress in decision making.

There will be occasions when the best answer will be to gather additional information or to improve the level of knowledgeon an asset or on a portfolio before embarking on a hierarchy of decisions. In this sense, decision-making is viewed as aseries of screening/scoping steps, likely with the goal of gaining insight, rather than complexity.

IOR/EOR screening techniques have been widely documented in the literature. Most of the screening techniques rely onconventional and advanced approaches or a combination of both (Guerillot, 1988; Joseph, et al. 1996; Henson, et al. 2002;Ibatullin, et al. 2002; Al-Bahar, et al. 2004). However, few studies focus on the decision making process from thedocumented screening studies. This paper briefly describes major steps of our screening methodology, well documented inthe literature, providing more details of various analytical and numerical simulation approaches used for different fieldstudies as part of the continuous development of the proposed IOR/EOR screening methodology. Additionally, case studiessummarized in this paper will also present the levels of information available and the decision made by the operator based onthe results of the screening. Field names and the operator are not disclosed for confidentiality reasons.

In addition to the introduction, a section on decision analysis attempts to explain some of the hurdles in the decisionmaking exercises to emphasize the need of framing and adequate level of modeling. A section on methodology summarizesflexible approaches developed by the authors, part of which has been documented in preceding papers. A number of decision-making exercises based on real field decisions follow. A discussion section and conclusions finalize the paper.

Decision Analysis ConsiderationsThe objective in this section is to show the basis for decision making in EOR/IOR. This section also illustrates some of thehurdles in decision making, particularly when EOR/IOR decisions are involved. A major part of this discussion will refer toframing of decisions. The Oil and Gas Industry presents its own peculiarities with respect to decision making (Mackie andWelsh, 2006).

A pressing issue in decision problems is framing. This appears simple at first glance, but understanding what the actualquestion is, avoids unnecessary complexity of the analysis and places the attention on the important issues, and not ontechniques and procedures. It is our proposition that the purpose of decision-making procedures is to provide the best state ofknowledge necessary for decision makers to arrive at rational decisions. Skinner (2000) clearly summarizes the concept ofFraming. This aspect of decision making goes hand-on-hand with Discovery, which includes the situation appraisal, placingthe decision makers objectives into a hierarchy and conducting an overall competitive analysis of the business situation, toreduce ambiguity. Framing helps to lower ambiguity with respect to goals or even eliminate conflicting objectives bydeveloping a decision hierarchy, strategy tables, and an influence diagram. Figure 1 shows an example of an influencediagram whose objective function is the Discounted Net Cash Flow and the decision is the well pattern, for a steam-flooding

project. Some of the value nodes are in reality chance nodes (uncertainties), but are not shown to simplify the diagram.A natural question is should we always rely on economic indicators, such as Net Present Value (NPV) to make a

decision? Pedersen, Hanssen and Aasheim (2006) discuss qualitative screening and soft issues, as they can set asidequantitative model recommendations. Ensuring that the model focuses on relevant decision criteria is a prerequisite for modelrelevance. The point to be made here is that NPV or other economic (hard) indicators should be used for hard, quantifiableissues, while a variety of methods can address soft issues, so that the balance between these two types of issues providesgood basis for decision alternatives. Some of the field cases discussed in later sections will illustrate this more clearly.

Bickel, Reidar and Bratvold (2007) present the results of a survey among decision-makers, support teams and academicsto try to understand the value of uncertainty quantification in decision making. A significant effort in the Oil and GasIndustry appears to have been dedicated to complex analyses of uncertainty quantification, perhaps in hope of eliminating it

or at least to reduce it. SPE, as a professional community, has organized a significant number of forums for uncertaintyevaluation and much fewer to decision-making. This might explain why such an intense focus has been place on uncertaintyanalysis as a goal in itself. One conclusion from the survey by Bickel et al. is that complexity of the decision analysis has notgreatly contributed to improving the decision-making process in our industry, at least as perceived by respondents to thesurvey. One interesting suggestion is that the decision analysis cycle is iterative, in the sense that if further assessments arerequired (or profitable data gathering), then the information should be gathered and the cycle repeated. In our opinion, adecision should be made, even if this signifies halting the project. One more significant bias is excessive use of intuition(expertise). Intuition can have a place in decision making (Dinnie, Fletcher and Finchm, 2002). However, this bias can hurtthe decision-making process in unexpected ways. The experience with chemical flooding a few decades ago led to theapparent definitive conclusion that these processes cannot be commercial. In this sense, this can serve as a screening criterionto discard chemical flooding operations. The difficulty is that new chemistry and process design have recently produced asignificant number of technical and economic successes by chemical flooding.


3/16

SPE 113269 3

Figure 1. Example of an influence diagram.

One important consideration in decision making in IOR/EOR is cognitive bias (Welsh, Bratvold and Begg, 2005). Thiscan take many forms, one of which reflects cognitive limitations of the human mind (Begg, Bratvold and Campbell, 2003).The level of risk aversion may not be consistent with goals, objectives and prudent decision-making. This is patently clearwhen value is destroyed because the decision-makers risk aversion is higher than that of the organization.

A number of methodological strategies have been developed over the years to deal with decision making for IOR/EOR projects. A particularly useful workflow for Decision and Risk Management is presented in Figure 2 (Goodyear and Gregory,

1994). In this workflow, screening based on critical variables of a number of IOR/EOR processes are used to determinefeasibility in a preliminary form. This step should not be performed prior to framing the problem, including some importantsoft issues, such as local availability of resources or even experience in IOR/EOR deployment. This sets the correct mentalstructures that simplify and guide our understanding of a complex reality (Bratvold and Begg, 2002). If screening does notyield feasible IOR/EOR opportunities, it might be necessary to stop the analysis. However, promising IOR/EOR strategiesmight be further screened by the use of prospective simulations, whether analytical or numerical on simplified sectors of thereservoir. Again, this second step may lead to detailed appraisal or to a full stop. Detailed appraisal is in part an exercise toreduce uncertainty and enable more complete economic evaluations as source of recommendation for decision makers. If theworkflow is successfully completed, then a project implementation follows.

Prospective Simulations

Screening

Detailed Appraisal

Project Implementation

Stop

Stop

Stop

Figure 2. Decision and Risk Management Workflow (adapted from Goodyear and Gregory, 1994).


4/16

4 SPE 113269

Our strategies for decision making mimic some of the steps described in Figure 2. However, we see this as a continuousexercise in screening and scoping to provide the best combination of soft and hard issues as inputs for decision makers. Inthis sense, it is often found that data gathering is one of the recommended courses of action. To mitigate some of thecognitive bias, data mining strategies are frequently used as part of advanced screening procedures. In this sense, instead ofrelying on a few experts biases, numerous biases are incorporated to the framed decision problem.

Methodology

Published results show that the combination of conventional and advanced IOR/EOR screening with fast analytical or smallscale numerical simulations represents a valuable approach for property acquisition and evaluation, to support of reservoirdevelopment plans (RDPs) and, most recently, to identify EOR opportunities in carbon sequestration.

The methodology implemented, though with variations from application to application, can be summarized as follows:1. Conventional screening based on a comprehensive comparison of reservoirs/fields under evaluation with an

extensive database of international IOR/EOR of approximately 2,000 projects is carried out. For this purpose, XY plots of average reservoir or fluid variable pairs are generated, e.g. oAPI gravity and depth, allowing engineers(analysts) to preliminary identify fields with similar properties. In this preliminary phase, engineers can infer lack offeasibility of various recovery methods to the reservoirs being analyzed by simply determining scarcity of fieldexperience for those methods in fields having reservoir properties similar to the fields under evaluation. Furtherqueries to the database using other variable pairs yield more closely paired combinations of potential fieldanalogues. This search for analogues can be seen as a data mining strategy in which the analyst looks at planes ofthe multidimensional space of all average variables used to represent the field cases. The choice of variables isguided by both availability of data and intuition (experience) of the analyst. Although desirable, a large collectionof reservoir and fluid variables is seldom available. What is meant by intuition is, for example, the accepted wisdomamong reservoir engineers that thermal processes are ideal for heavy-oil reservoirs. Analogue searches can providegood analogues in light-oil reservoirs, for which steamflooding might look like a good option. However, thisapproach does not necessarily help to identify technical success of a recovery process or lack thereof. Radar plots ofup to six variables are also used to identify trends and ranges of preferences for (or the applicability of) a particularEOR method in multiple reservoirs prior to starting the use of advanced screening methods (Manrique and Pereira,2007). In the case of commercially unproven EOR processes (i.e. THAI, VAPEX, etc.), this screening phaseanalyzes and estimates technical feasibility of these recovery methods based on a more theoretical and engineering

judgment basis given the lack of production data and field experiences. The notions that are conveyed bycomparison with abundant field cases database is the level of risk and bias containment, so that wisdom does not

become an excessive cognitive bias.2. In a second phase, IOR/EOR conventional screening is complemented with the use of commercial analytical tools to

expand the evaluation and hence further validate the applicability (feasibility) of the most feasible recovery processin the field under evaluation. Analytical screening with commercial tools is also based on the comparison ofreservoir properties of the field under evaluation with property intervals of known IOR/EOR projects existing in thedatabases. Two known screening procedures are used to expand on screening. The first procedure is based on go/nogo criteria (ARC, 2006), while the second one uses fuzzy logic to generate scores for ranking, based on triangulardistribution of comfort intervals (IRIS, 2007). Since the bias (knowledge base or expertise bias) differs in the twoscreening procedures, these additional screening approaches provide a broader evaluation of the property of interest.This is not unlike asking for opinion to different experts on the feasibility of EOR/IOR opportunities for an asset.

3. This step relies on geological screening. The first phase of this type of screening is to compare IOR/EOR projectsavailable in the database with the field under evaluation on the basis of geologic characteristics, e.g. trap type,depositional environment, lithology, type of structure, diagenesis. This first evaluation helps to identify whether a

particular IOR/EOR process has been implemented in a particular geologic formation. This represents a qualitativeor soft input variable as described here, but it is a different important set of features that impact IOR/EOR processes.In the case of sandstone reservoirs, this analysis is augmented by using the matrix of depositional environment vs.

lateral and vertical heterogeneities documented in the literature (Tyler and Finley, 1991; Henson, Todd and Corbett,2002). Although the location of EOR projects as a function of the depositional system heterogeneities is somewhatsubjective, due to the lack of geologic information and/or differences in the geologic interpretation, we still believethat this type of analysis can guide the decision-making process associated with EOR projects based on previousfield experiences. If the dimensions of sand bodies or genetic units (length, thickness, and width) and current or

proposed well length and spacing are known, horizontal and vertical heterogeneities indexes can be estimatedthrough simple equations (Henson, Todd and Corbett, 2002). This analysis has been successfully applied to estimateSAGD well pair horizontal length, vertical separation and spacing in Canadian Oil Sands (Manrique and Pereira,2007). Additionally, 2-D and 3-D heterogeneity index analysis have been proven to be more robust when combinedwith Dykstra-Parson (DP) coefficients calculated from well log (petrophysical analysis) and core permeability data.DP coefficients also have been used extensively to generate full field maps as a quick quality control during fullfield petrophysical studies (i.e. impact of petrophysical cut-offs on reservoir heterogeneity) as part of detailedreservoir (Integrated) engineering studies. DP maps combined with other reservoir properties, i.e. net pays and


5/16

SPE 113269 5

fluid saturation, have been also used in full field analytical simulations to evaluate RDP under EOR processes(Alvarado et al., 2003). In field case studies where only a permeability interval is available, DP maps can begenerated to estimate the impact of reservoir (lateral and vertical) heterogeneities on oil recovery or CO 2 plumemigration in saline aquifers, among other applications.

4. IOR/EOR conventional and geological screening is followed by advanced screening techniques based on artificialintelligence, data mining and space reduction techniques well documented in the literature (Alvarado et al., 2002;Manrique, Alvarado and Ranson, 2003). To perform this advanced screening, mined data from roughly 450

successful IOR/EOR projects are compared with the reservoir(s) under evaluation. The simultaneous projection of areduced set of reservoir variables, namely temperature, reservoir depth, current reservoir pressure, porosity, permeability, API gravity, and viscosity, is represented on 2-D maps (Expert Maps). Clusters in these 2-D projections represent different reservoir types (Reservoir Typology). Experience shows that the 2-Drepresentations of reservoir clusters have in common the types EOR projects implemented. Multidimensional

projections on the 2-D plots lead to the simultaneous comparison of multiple variables and, more importantly, aconvenient clustering of reservoir types. This way, in addition to obtaining analog cases, statistics on recoveryfactors can be obtained, adding robustness and reducing Expert Bias to the screening evaluation. This screeningstage can be thought of as a scoping strategy, because numerical indicators such as recovery factor that clearlyimpact economics in projects can be obtained readily. This advanced screening has been also developed specificallyfor CO 2 injection for CO2-EOR and sequestration evaluations (Manrique, Alvarado and Ranson, 2003; Velasquez,Rey and Manrique, 2006).

5. Another important step of IOR/EOR screening methods should include the evaluation of soft variables that needto be incorporated early in the evaluation to avoid spending time considering IOR/EOR processes that may not befeasible due, for instance, to the lack of injection fluids and environmental constraints, among others. Because someof these decisions are ongoing technology developments and not entirely proven, they are not amenable to same typeof hard analysis based on economic indicators. Some examples of the analysis and discussions included in thisstep of the evaluation are:

a. CO 2-EOR and CO 2 capture and sequestration have attracted increasing attention in the U.S. and abroad.However, despite the applicability of CO 2 floods in a particular reservoir, fields under evaluation frequentlydo not have available CO 2 within reasonable distances. Due to the high cost of CO 2 capture andcompression from power generation plants, the preferred choice is the use of natural sources of CO 2.However, this may reduce the feasibility of CO 2 flooding in a particular field due to the cost associated tothese projects (i.e. CO 2 cost, pipeline construction, OPEX, etc.). In other words, the required CAPEXmight be too high for the incremental oil recovery expected from the field. In this type of evaluation, GIStechnology represents a valuable tool to map all CO 2 sources (natural and anthropogenic) in the vicinity ofthe field under evaluation.

b. Current and future trends in oil prices have created a remarkable interest in heavy and extra heavy oil property evaluation and acquisitions in the U.S. The obvious options to develop these resources are EORthermal methods, especially steam injection. However, it is important to consider the options available togenerate the required steam, e.g. natural gas or bitumen gasification, at early stages of evaluation, giventheir impact on EOR project timing and economics. Another option to be considered for development ofheavy oil resources is the use of down-hole heating technologies combined with water injection. Althoughthis option has not been fully tested, it represents a potential advantage because surface facilities and heatlosses are minimized and projects can start earlier because steam generation plants are not required.

6. Once the most feasible IOR/EOR processes are identified, the next phase of the evaluation includes analytical and/ornumerical simulation, depending on the amount and quality of data as well as time constraints. Based on thedecision framework, the screening study is defined (what needs to be answered or the decision involved in thescreening study), performance prediction using analytical and numerical simulations or both is defined at earlystages of the evaluation. It is well known that numerical reservoir simulation plays a key role in the development of

reservoir management plans, reservoir monitoring, and the evaluation of reservoir performance under different EORmethods. It is also well known that numerical simulation studies are costly and time consuming in addition torequiring highly trained professionals capable of simulating and understanding of the physics behind EOR

processes. Although the present methodology does not pretend to replace numerical simulation with analyticalsimulations, there are cases where full numerical reservoir simulation studies are not justified with the available dataand/or time constraints. It is a fact that oil production forecasts obtained from analytical simulations tend to beoverly optimistic, given the simplicity of the approaches used in estimating oil recoveries. However, for fastscreening purposes, analytical simulations provide key insights, sensible parameters, and help to identify theuncertainties associated with different recovery processes. Additionally, preliminary economic evaluations can berun using these optimistic production profiles, along with varying oil prices, for the purposes of screening. If

projects do not have economic merits using the optimistic production profiles obtained from analytical simulations,most certainly the project economics will be less attractive when proper simulation studies are done. The section offield cases in this paper will focus on the description of different analytical, e.g. RDP, optimum well spacing, and 2-


6/16

6 SPE 113269

D and 3-D numerical, namely conceptual models, oil-sand development plans, up-scaled sector models, CO 2 plumemigration and fully compositional element of symmetry, among others, simulation strategies used for differentdecision management strategies based on the proposed methodology.

7. The final phase of the IOR/EOR screening methodology is the economic evaluation of most feasible IOR/EOR processes. Simple economic models and indicators are used to rank IOR/EOR options as part of the screening anddecision analysis process. However, it is a reality that economic evaluations may vary from company to company.Therefore, the operator or investor generally takes over this step, unless the supporting team and decision-makers

are part of the same organization. The economic calculations can be linked to simulations and decision risk analysiseither by exporting all possible production profiles to commercial software, or by developing the required interfacesin Excel or Visual Basic. In these cases, the economics are run considering the main input criteria andconstraints provided by the operator or investor (decision maker). The economic evaluation can also be run usingoptimistic production profiles and different well costs and oil prices for the purposes of screening. If projects areunprofitable or too sensitive under the conditions evaluated, they can be discarded and re-examined during the next

planning period. However, later examination will depend on the risk tolerance of the operator or a particularinvestor for a specific EOR technology. What is important is that an oil reservoir needs to be re-evaluated

periodically to determine how current development plans may impact EOR processes in later stages of production because of changes in reservoir conditions. Potential oil recovery is dynamic and changes as a reservoir matures andas reservoir energy evolves.

Besides the importance of IOR/EOR screening studies documented in the literature, the decision made from this analysisis also important but have not been fully documented in the literature. Following section will briefly describe different fieldcases evaluated in the U.S. and abroad describing also the decisions made based on the results of this fast screening analysisand evaluation.

Field CasesSeveral cases are presented in this section to illustrate a variety of decision-support exercises. All cases refer to real businessdecisions in the context of IOR/EOR, but names and locations are not disclosed to protect confidentiality of the sources.

Field cases presented in this paper include shallow (500 ft) to deep (15,000 ft) reservoirs with oil gravity ranging from7API (Canadian Oil Sands) to gas condensate reservoirs (60API) in South America. Fields evaluated include sandstone(consolidated and unconsolidated) and carbonate reservoirs, among other lithologies.

To illustrate the different types of decisions, context and constrains of the decision-making problems, cases were dividedaccording to the availability of data and time constraints for the decision-making process.

Field Cases Type I: Lack of data and time constraints.IOR/EOR screening studies lacking data and having stringent time constraints for the decision-making process (typically

from few weeks to up to 3 months) are the most common decision-support cases developed in recent years. In terms of theworkflow of our methodology, case studies cover mostly steps 1 thru 5 described above. However, the details of eachscreening study vary as a function of the information available and the decisions to be made by the investor or the operator

based on the results obtained. The proposed screening methodology has been used to evaluate technical feasibility ofIOR/EOR processes in Asia, Canada, Central America, South America and the U.S. Common decisions or questions thatneed to be answered from IOR/EOR screening studies can be exemplified with the following list:

Determine the most feasible IOR/EOR processes including preliminary analytical simulations to estimate oilrecovery potential

Justify data gathering programs, i.e. drill and log wells, core recovery and fluid samples, etc. Justify detailed laboratory studies, i.e. chemical waterflooding, minimum miscibility tests, special core analysis, etc. Justify more detailed engineering (Phase II) studies Generate preliminary RDP based on one or more EOR process, among others.

Some examples describing the EOR decision-making procedures follow in the subsections.

Case study A.Field case A was posed as a screening problem, including preliminary analytical simulations, of a light-oil dolomitic

formation in the U.S. The main objective of this screening study developed in early 2007 was to identify the most feasiblerecovery processes and identify potential closer analogs to define a more detailed simulation study and potentially a pilot test.Results from the screening study showed that gas injection (continuous injection or in a WAG mode) and waterflooding to bethe most viable IOR/EOR options for this field. Se veral reservoir analogues in which N 2 (e.g. Binger Field, OK) or CO 2 (e.g.Charlson, Kutler and University Wadell) had been injected were reported in the literature (see Figure 3). Additionally,waterflooding projects in low permeability (< 1.5 md) dolomite reservoirs were also identified in some Texas fields (e.g.Levelland, Mabee and Robertson North) after several queries to the database. Potential field (closer) analogues identifiedhelp to increase the operators level of confidence associated with the development of the field. Although preliminaryanalytical performance prediction showed that gas flooding (continuous or in WAG mode) outperformed waterflooding, water


7/16

SPE 113269 7

injectivity tests were recommended to validate a WAG injection strategy due to the low average permeability of the field. Giventhe applicability of N 2 and CO 2 (Flue Gas) flooding in Field A, high pressure air injection (HPAI) was not discarded at thislevel of evaluation. HPAI was considered an option after analyzing the availability of CO 2 (anthropogenic and natural) sourcesnear the field and the estimated N 2 MMP (from empirical correlations) as well as the costs associated to air separation and N 2 rejection units. Results of the study justified a data gathering program to develop the required laboratory and field data to identify

potential pilot areas in the field for the next phase of the evaluation.

CO2 Misc.

N2 Immisc.

N2 Misc.

Polymer

WAG-CO2 Misc.

WAG-HC Misc.

air

Water flooding

Field A - Case 1

Field A - Case 2

Field A - Case 3

Cluster 2-3Method %

N2 Immiscible 50.0Polymer 25.0Water Flooding 25.0

Cluster 2-4

Method %

WAG-HC 38.5

N2 Miscible 23.1

Water Flooding 23.1

N2 Immiscible 15.4

Cluster 2-1Method %

Water Flooding 47.6WAG-CO 2 Misc. 23.8

Air 9.5CO 2 Misc. 9.5Polymer 9.5

Cluster 2-2Method %

Water Flooding 42.9

CO 2 Misc. 14.3N2 Miscible 14.3WAG-CO 2 Misc. 14.3

WAG-HC 14.3

Binger

East Binger

Field A sensitivity cases

CO2 Misc.

N2 Immisc.

N2 Misc.

Polymer

WAG-CO2 Misc.

WAG-HC Misc.

air

Water flooding

Field A - Case 1

Field A - Case 2

Field A - Case 3

Cluster 2-3Method %

N2 Immiscible 50.0Polymer 25.0Water Floodin g 25.0

Cluster 2-4

Method %

WAG-HC 38.5

N2 Miscible 23.1

Water Flooding 23.1

N2 Immiscible 15.4

Cluster 2-1Method %

Water Flooding 47.6WAG-CO 2 Misc. 23.8

Air 9.5CO 2 Misc. 9.5Polymer 9.5

Cluster 2-2Method %

Water Flooding 42.9

CO 2 Misc. 14.3N2 Miscible 14.3WAG-CO 2 Misc. 14.3

WAG-HC 14.3

Binger

East Binger

Field A sensitivity cases

Figure 3. Expert map for Case A advanced screening.

Case study B.Field case B included the screening and evaluation of a Canadian Oil Sand property. The main question in this screening

study was to identify whether the property could be developed by Steam-Assisted-Gravity-Drainage (SAGD) afterunsuccessful cyclic steam injection pilot tests in the late 1980s. The study was divided into two phases; the first one was thescreening study that included preliminary analytical simulations and a second phase that included a 2-D simulation studyusing an existent geologic interpretation of the property under evaluation. The methodology followed for this field case andseveral other Canadian Oil Sands has been recently published (Manrique and Pereira, 2007).

Results of this screening study showed that both SAGD and steam injection (cyclic or continuous steam) are applicable toField B. Potential field analogues were identified for SAGD, i.e. Burnt Lake, and cyclic steam injection, i.e. Cold Lake andPeace River. However, analytical simulations consistently showed that SAGD outperformed cyclic steam injection in a

broad range of reservoir and steam injection conditions. Once the geologic interpretation became available we noticed thatField B showed lack of continuous net pay areas of 20 m or thicker limiting the applicability of SAGD in the property

based on conventional screening criteria well documented in the literature (Butler, 1991; Palmgreen and Renard 1995; ARC,2006). Additionally, the geologic interpretation also showed that thicker areas ( 20m) were not always continuous due tothe presence of interbedded thin layers of low permeability that may negatively impact the steam chamber development and

hence cumulative steam-oil ratios (CSOR) and oil recovery factors. To validate our findings, a 2-D parametric numericalsimulation study was carried out to estimate the potential of SAGD in Field B. In the absence of sufficient reservoir data torun a proper numerical study, PVT and relative permeability data were estimated from public documents. Sensitivities on net

pay, presence of top gas and bottom water were also run to estimate the impact of these critical variables on cumulative oil production and CSOR. The 2-D parametric numerical simulation study helped to identified areas where SAGD wasapplicable. Additionally, detailed review of successful cyclic steam injection projects reported in Cold Lake, Peace Riverand Primrose justified a detailed review of past experience of this recovery method in the field. Finally, the screening studyidentified documented potential development plan schemes under SAGD and cyclic steam injection that currently are underconsideration by the owner of the field.

Case study C.The decision problem associated with Field Case C was the well spacing for a steam flood project with a short time frame

for making the decision. The analysis was cast as a decision problem having efficient recovery/production as the objective


8/16

8 SPE 113269

function. Since building a representative petrophysical model for numerical simulation was not feasible given the timeconstrain and available data, the question was: how to use the best of the available data to provide a meaningfulrecommendation in the given period of time?

In these situations when large volumes of data are obviously available, it may be tempting to argue that a detailed modelis the solution. However, in this case, the time allocated was insufficient for detailed numerical simulation and key data tocomplete a detailed reservoir model were not available. Our approach is to integrate the available data and use analyticaltools and some small conceptual models. This is not unlike moving mosaic techniques used for infill drilling (Voneiff and

Cipolla, 1996; Guan et al., 2002; Hudson, Jochen and Jochen, 2000; Hudson, Jochen and Spivey, 2001; Wozinak, Wing andSchrider, 1997) or the quality map approach (Da Cruz, Horne, and Deutsch, 2004). A frequent approach to analyticalsimulation consists of modeling a well pattern (generally 5-spot patterns), with the assumption that the obtained resultsrepresent the average performance of a particular sector of the reservoir. Given that this assumption is strong, as properties ina reservoir should be statistically equivalent to be valid, care should be interpret this type of evaluation. For this purpose, theconceptual or analytical modeling accounts for variability in the reservoir by evaluating the performance of each pattern withareal distribution on the reservoir map, on statistically equivalent basis and the distribution of critical parameters for oilrecovery processes such as Dykstra-Parson (DP) coefficient or some other heterogeneity indexes . The DP coefficient wasused in Case C as a heterogeneity index to generate the quality map for each reservoir as shown in Figure 4 .

Several different quality maps have been obtained per reservoir, for example the DP or net pay maps, or a combination of both indicators. Although some other parameter could have been used as a quality map, here DP and net pay maps were theonly ones used. Hundreds of analytical simulations were run for each quality map. For example the quality map shown inFigure 4 is divided into different heterogeneity regions (based on DP differences), and for each heterogeneity region severalanalytical simulation runs were completed accounting for well spacing and samples of the available data. The recovery factorfor each pattern for different well spacing was calculated and then the values of the recovery factor were weighted by thevalue of the original oil in place. The recovery factor is shown in Figure 5 for different well spacing values. Besides the wellspacing issues, several other sensitivity parameters such as steam quality were studied. Figure 6 compares the recoveryfactor for different steam quality and well spacing after 0.7 pore volume of steam injected.

The production and injection forecasts were transferred to a simple economic model to calculate net present value foreach scenario. The economic model was extremely simple, so rather than focusing on the accuracy of the NPV, attention was

placed on the relative economic performance of the different scenarios. Two oil price cases ($60/bbl and $90/bbl) were run.In any infill case a number of wells had to be drilled. The drilling and completion costs were estimated at $1,000,000 perwell for both injectors and producers. This cost includes surface facilities and the cost of steam was assumed $1 per barrel.

Net cash flow was calculated and the results are depicted in Figure 7 . These results show that the optimum well spacing is150m. We will return to the a posteriori interpretation of this evaluation in the section of this paper, to compare withnumerical simulation results.

5000 100005000 10000

Figure 4. Dykstra-Parson coefficient map for one of the reservoir in Field C


9/16

SPE 113269 9

0

10

20

30

40

50

60

0 20 40 60 80 100

% Pore Volume Injected

R e c o v e r y

F a c

t o r ,

%

Well Spacing = 75m

Well Spacing = 150 m



Figure 5. The recovery factor for one pattern among different well spacing

0

10

20

30

40

50

0 50 100 150 200 250 300

Well Spacing (m)

R e c o v

e r y

F a c

t o r ,

%

Steam Quality=1.0

Steam Quality=0.8

Steam Quality=0.6

Steam Quality=0.4

Steam Quality=0.2

Steam Quality=0.0

Figure 6. Steam quality sensitivity among different well spacing


10/16

10 SPE 113269

$0

$200

$400

$600

$800

$1,000

$1,200

$1,400

$1,600

$1,800

0 50 100 150 200 250 300 350

Well Spacing, (m)

N e

t P r e s e n

t V a l u e

PVI=70%, Oil Price = $90

PVI=70%, Oil Price = $60

Figure 7. Net preset value at 70 % of pore volume injected with two different oil prices.

Field Cases II: Not enough time to use sufficient data.In this type of case field studies the volumes of data may be sufficient to build detailed models for which all levels of

screening can be completed, but stringent time constraints (from 2 to 4 months) make IOR/EOR screening studies morechallenging because the decision-making process is more complex. In these field case studies all steps of the proposedmethodology can be covered. However, most of the time the changes and challenges are related to the simulation approachused to help the investor/operator make a decision. As it has been mentioned earlier, each screening study is specificdepending on the information available, geographic location and access to oil and gas markets and investor/operator decisionframework, among others. Common decisions or questions answered from recent IOR/EOR screening studies are:

Initiate visualization studies to estimate most feasible recovery processes and RPD as part of Front-End-Loading(FEL) studies

Evaluate EOR technologies and potential implementation strategies of one or a portfolio of reservoirs Analyze reservoir portfolios to evaluate CO 2 EOR and sequestration potential Justify large investment decisions to develop pilot tests or comprehensive data gathering programs associated to

EOR projects Justify more detailed engineering (Phase III to V) studies (i.e. full field simulations studies, EOR project design and

monitoring, investments in surface facilities, etc.) Complete property evaluation and acquisitions

Some examples of this type of decision-making process developed in Western Hemisphere countries using the proposedmethodology are briefly described here.

Case study D.Field case D is part of a visualization study to identify potential EOR technologies and RPD in a portfolio of gas

condensate and light crude oils reservoirs (11 multi reservoir fields) in South America. The identification of technologies,recovery processes and production strategies to maximize gas and condensate production represents a key objective of theanalysis. Ranking most feasible options to maximize gas and condensate recovery was also a key objective of this phase ofthe analysis. After the evaluation of hydrocarbon gas in place, stochastic volumes (P10, P50 and P90) were used to generate

production profiles and cumulative recoveries considering different production scenarios. Gas and condensate production predictions were based on analytical approach (development planning excel spreadsheet). The model used assumes that thegas field obeys a P/Z vs. cumulative gas produced relationship. Despite the simplicity of the model, it is a rather rigorous

program to run which includes a drilling program, inflow performance information, gas injection and/or re-injection andsurface compression, among other activities. Typical production profiles calculated from this evaluation and used ineconomic calculations is shown in Figure 8. Cumulative production information is depicted in Figure 9.

Regarding EOR opportunities, N 2 injection was found to be the optimum recovery strategy to increase gas and especiallycondensate recovery. Additionally, N 2 injection will contribute to maximize gas sales by reducing dry gas re-injection.However, gas recovery is not as efficient because of the higher capital and operating expenses to separate N 2 production from


11/16

SPE 113269 11

produced gas streams (N 2 rejection units). Sales gas also may be negatively impacted if early N 2 breakthrough isexperienced, especially in some of the eolian formations existing in the area of evaluation. Finally, preliminary economicssuggests that other scenarios evaluated are outperforming N 2 injection, especially due to the high CAPEX and OPEX of N 2 injection projects. Therefore, at this stage of the evaluation it was possible to identify that even in the best case scenario of

N2 injection the economics do not show enough merit to justify further analysis of this scenario. This demonstrates the valueof the proposed fast screening methodology by helping the operator to focus on most valuable options identified.

0.E+00

1.E+03

2.E+03

3.E+03

4.E+03

5.E+03

6.E+03

7.E+03

8.E+03

2009 2014 2019 2024 2029 2034 2039

Year

M b b l p e r y e a r

0

20

40

60

80

100

120

B s c

f p e r y e a r

Condensate Prod. Rate - All Fields

Cond. Prod. Rate - Field D1

Gas Prod. Rates - All Fields

Gas Prod. Rate - Field D1

Figure 8. Yearly condensate and gas production for field D1 and the total for all fields under N 2 injection.

0.E+00

1.E+04

2.E+04

3.E+04

4.E+04

5.E+04

6.E+04

2009 2014 2019 2024 2029 2034 2039

Year

C u m .

C o n

d e n s a

t e P r o

d u c

t i o n

( M b b l s )

0

200

400

600

800

1000

1200

1400

1600

1800

2000

C u m .

G a s

P r o

d u c

t i o n

( B s c

f )

Cum. Condensate Prod. - All Fields

Cum. Condensate Prod. - Field D1

Cum. Gas Prod. - All Fields

Cum. Gas Prod. - Field D1

Figure 9. Cumulative condensate and gas production for Case D (Field D1 and All fields) under N 2 injection.

Case study E.This field case represents an evaluation of potential CO 2-EOR, CO 2 sequestration and CO 2 market opportunities in a

portfolio of oil reservoirs. The identification of gas fields and saline aquifers as potential CO 2 sequestration options was alsoincluded in this evaluation. However, this section will be focused mainly on the CO 2-EOR opportunities. Approximately100 oil reservoirs were identified within 100 square miles from the location of a planned coal-fired power plant. It isimportant to mention that the volume of data to process was high due to the number of fields identified. However, in many


12/16

12 SPE 113269

of the fields identified showed scarce data. In these cases, input data were generated from correlations of closer fields producing from the same formations. To meet project goals and timeline (approximately 4 months) the analysis focused onidentifying potential CO 2 markets for EOR within a small-scale area (50 sq. miles) and targeting only oil fields with an OOIPgreater than 10 million barrels, since smaller fields do not add large EOR or CO 2 sequestration potential (Figure 10). Takingthese constraints into account, it was assumed that if a CO 2 market is justified, it may expand with time once CO 2 distribution

pipelines are in place.

Figure 10. Plant site location and possible candidates for CO 2 EOR and storage.

A second filter to scope fields for CO 2-EOR was based on current reservoir pressure, oil viscosity and most importantlycurrent production rates, if available. The purpose of this second fast ranking procedure was to identify fields that showhigher probabilities of CO 2 miscible processes and potentially higher recovery factors (large EOR potential). Based ondetailed screening and analytical simulations, a few candidates for CO 2-EOR were identified. Key findings of this screeningevaluation indicated that there was a lack of CO 2-EOR opportunities in the vicinity of the planned coal-fired power plantequipped with CO 2 capture capabilities. Additionally, by the time the coal-fired were built, it would be highly probable thatfield candidates might not be in operation and therefore justify large capital investments associated for CO 2-EOR projects. Insummary, the conclusion of this evaluation strongly suggested that sequestration opportunities in the area under study shouldfocus on saline aquifers given the volumes expected to be sequestered and once proper regulatory framework (i.e. well

permitting, CO 2 liability, etc.) is in place.Based on the results of the screening study, a second phase of the project was developed to identify major geologic

structures that can store large volumes of CO 2 for geologic time periods. Phase II of the project identified four potentialsaline aquifers to store CO 2. Each of the cases was simulated numerically assuming different scenarios to estimate number ofwells required and plume migration after 300 yr for different volumes of CO 2 captured/injected. However, deep salineaquifers identified did not have enough well data to develop an adequate reservoir description. Therefore, the impact ofreservoir heterogeneity on CO 2 injectivity, storability and plume migration was estimated through randomly DP coefficients(homogeneous, base and heterogeneous cases) distributions using average rock properties and different net pays. Thesimulation approach contributed to preliminary rank CO 2 sink options identified providing potential development plans andvaluable information (i.e. number of injectors, plume migration vs. land ownership, data gathering and monitoring programs,etc.) that has contributed to the decision-management strategy to design and prepare next phase of the project.

Case study F.Field case F consisted in the screening and evaluation that supported the decision-making process of a property

acquisition during the second half of 2006. The property under evaluation is a high pressure light oil carbonate reservoir in


13/16

SPE 113269 13

the U.S. in operation since the 1970s. In this case study detailed screening analysis was carried out to identifying mid- andlong-term upside potential for the field. However, the most challenging aspect of this study was to develop a reasonablenumerical simulation study to evaluate multiple production optimization strategies in a short period of time (less than 3months). Neither a full field geological model, nor a numerical model was available at the time of the evaluation. However,a large volume of good quality data (i.e. historical production, core data, multiple reports, etc.) was available to be used todevelop the numerical simulation study. The later represented a difficulty give the timeframe to make the decision ofwhether or not to acquire the field under evaluation. Therefore, the decision-making process was somehow one of the most

challenging steps of this study.Give the amount of data, budget and time constraints, different compositional numerical simulation (history matching and prediction) approaches were discussed. Full field and large sector models were discarded due to budget and time constraints.The selection of specific well patterns was also discarded avoiding the selection of an area (i.e. too good or too bad well

patterns) not necessarily representative of the entire field. The final decision was to generate a well pattern (Element ofsymmetry) using average reservoir properties and production/injection history of the field and based on detailed reportsdocumented by the operator (seller). This strategy was considered the best option to deliver the results in a timely manner.

The element of symmetry represented approximately 1% of the total reservoir volume. History match and performance prediction were run using a 9 pseudo-component compositional model. Results were up-scaled to the OOIP of the field providing expected recovery factor values and injection and production profiles for different scenarios. Results weredelivered to the client to run the economics. Based on the results and multiple decision-risk management sessions the fieldwas purchased. After a year the field has been performing in reasonably good agreement with the simulation results andadditional upside potential has been identified from more detailed studies following an integrated reservoir analysis approach.

Does model complexity always affect decisions?One interesting question is to what extent the accuracy of a model affects the outcome of decision-making. The answerdepends on the project context, but some decision makers often distrust simplified modeling strategies due to the lack ofaccuracy. This is the case of analytical simulation and conceptual sector models for IOR/EOR. Given limitations anduncertainties in data sources (models) and time constraints, simplified analyses are often a better solution to certain decision-making problems. Embarking on IOR/EOR projects, with a hierarchy of successive interdependent decisions can facilitatethe task of promoting further development in a field.

To try to partially answer the question regarding the influence of model complexity, Case C is recalled here. Analyticalsimulation relied on layer-cake models, for which petrophysical data from interpreted well logs were used to assigned DPcoefficients. This representation of heterogeneity added to the limitations of analytical simulation can be, in principle,considered worthless from the decision-making point of view. A comparison was drawn with numerical simulation models.The distribution of permeability of the numerical model reproduced the value of the DP coefficient, but as opposed to theanalytical models, permeability was distributed isotropically in 3D (Figure 11). Besides the difference in the solution

procedure, the model in this case offers more degrees of freedom than the analytical cases.The comparisons between numerical models and analytical model are presented in terms of recovery factor versus

percentage of pore volume injected. Steam flooding results are presented in Figure 12, for two of the well spacing valuesconsidered . In steam flooding higher well spacing results in higher heat lost, therefore, for the same pore volume injected,recovery factor drops in the higher well spacing. Numerical models show that the results from analytical models areoptimistic for different well spacing.

Despite the significant differences in production profiles (represented here by the value of the recovery factor), a post-mortem analysis using numerical simulation results showed that the final recommendation was not affected by the simulation

procedure. Although this will not always be correct or true, what turns out to be true is that even if a complete andsophisticated model were available, the reality of production can be badly predicted.


14/16

14 SPE 113269

DP=0.1DP=0.5DP=0.3

DP=1.0DP=0.9DP=0.7

DP=0.1DP=0.5DP=0.3

DP=1.0DP=0.9DP=0.7

Figure 11. Permeability distributions for numerical models for Field Case C.

0

10

20

30

40

50

60

0 20 40 60 80 100

% Pore Volume Injected

R e c o v e r y

F a c

t o r ,

%

Well Spacing = 75 m, Analytical

Well Spacing = 75 m, Numerical

Well Spacing = 300 m, Analytical

Well Spacing = 300 m, Numerical

Figure 12. Comparison between Analytical models and numerical for four different well spacing

Conclusions

Field cases briefly described in this paper demonstrate that decisions can and have been made without the necessity ofsophisticated techniques and time consuming studies. Proper engineering judgment and physically sound analysis representkey variables in this type of evaluation. Finally, it is important to mention that the authors do not propose to substituteconventional and rigorous numerical simulation and detailed reservoir studies. However, it is also important to remind thatmanagement decisions and business opportunities cannot necessarily afford to wait for all the data or comprehensive studieswe all engineers like or wish to have. Moreover, issues often attributed to uncertainty could be dealt within the realm offraming. Ambiguity is more often the cause of the over-analysis trap, than the actual complexity of the contextualizeddecision problem.

Bias in the screening exercise can be mitigated by the use of several screening procedures, including those that rely ondata mining. These procedures incorporate as much as possible, experience-based guidance for IOR/EOR screening.Decision-making questions posed with the use of unproven technologies are good examples for which soft issues can play asignificant role.


15/16

SPE 113269 15

AcknowledgementsThe authors would like to thank Norwest Questa Engineering for permission to publish this paper. The authors are indebtedto Patrick Hahn, Tom Farris, Carlos Pereira, and Anibal Araya for their contributions to simulation efforts and data analysis.Thanks go to John Wright for critically reviewing the manuscripts and for suggestions.

ReferencesAl-Bahar, M. A., Merril, R., Peake, W. and Reza Oskui, M. J. 2004. Evaluation of IOR Potential within Kuwait. Paper SPE 88716

presented at the 11th

Abu Dhabi International Petroleum Exhibition and Conference, Abu Dhabi, U.A.E., October 10-13, 2004.Alvarado, V., Ranson, A., Hernandez, K., Manrique, E., Matheus, J., Prosperi, N. and Liscano, T. 2002. Selection of EOR/IOROpportunities Based on Machine Learning. Paper SPE 78332 presented at the 13 th SPE European Petroleum Conference, Aberdeen,Oct. 29-31, 2002.

Alvarado, V.; Manrique, E.; Vasquez, Y., Noreide, M. 2003. An approach for full-field EOR simulations based on fast evaluation tools.Paper presented at the 24 th Annual Workshop and Symposium for the IEA Collaborative Project on Enhanced Oil Recovery. Canada,September 7-10, 2003.

Begg, S.H. and Bratvold, R.B. 2003. Shrinks Or Quants: Who Will Improve Decision-Making. Paper SPE 84238 presented at the SPEAnnual Technical Conference and Exhibition, Denver, Colorado. 5-8 October 2003.

Bickel, J.E. and Bratvold, R.B. 2007. Decision Making in the Oil and Gas Industry: From Blissful Ignorance to Uncertainty-InducedConfusion. Paper SPE 109610. Paper SPE 109610 presented at the 2007 SPE Annual Technical Conference and Exhibition, Anaheim,California, 11-14 November 2007.

Bratvold, R.B. and Begg, S.H. 2002. Would You Know a Good Decision if You Saw One? Paper SPE 77509 presented at the SPE AnnualTechnical Conference and Exhibition, San Antonio, Texas, 29 September-2 October 2002.

Butler, R. M. 1991. Thermal Recovery of Oil and Bitumen, Prentice Hall, Englewood Cliffs, New Jersey, 1991.

Dinnie, N.C., Fletcher, A.J.P., and Finch, J.H. 2002. Strategic Decision Making in the Upstream Oil and Gas Industry: Exploring Intuitionand Analysis. Paper SPE 77910 presented at the SPE Asia Pacific Oil and Gas Conference and Exhibition, Melbourne, Australia, 8-10October 2002.

Dubois J.R. 2004. Portfolio Management Using Questionable Quality Data. Paper SPE 90395 presented at the SPE Annual TechnicalConference and Exhibition, Houston, Texas, 26-29 September 2004.

Da Cruz, P.S., Horne, R.N., and Deutsch, C.V. 2004. The Quality Map: A Tool for Reservoir Uncertainty Quantification and DecisionMaking. SPEREE (February 2004). SPE-87642.

Fletcher, A. and Davis, J.P. 2002. Decision-Making with Incomplete Evidence. Paper SPE 77914 presented at the SPE Asia Pacific Oil andGas Conference and Exhibition, Melbourne, Australia, 8-10 October 2002.

Gerbacia, W.E. and Al-Shammari, H. 2001. Multi-Criteria Decision Making in Strategic Reservoir Planning Using the Analytic HierarchyProcess. Paper SPE 71413 presented at the 2001 SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana 30September-3 October 2001.

Goodyear S. G. and Gregory A. T. 1994. Risk Assessment and Management in IOR Projects. Paper SPE 28844 presented at the 1994European Petroleum Conference, London, October 25-27, 1994.

Guan L., McVay D. A., Jensen J. L., and Voneiff G. W. 2002. Evaluation of a Statistical Infill Candidate Selection Technique. Paper SPE75718-MS presented at the SPE Gas Technology Symposium, 30 April-2 May, Calgary, Alberta, Canada, 2002.

Guerillot, D. R. 1998. EOR Screening With an Expert System. Paper SPE 17791 presented at the Petroleum Computer Conference, SanJose, California, June 27-29, 1988.

Henson, R., Todd, A. and Corbett, P. 2002. Geologically Based Screening Criteria for Improved Oil Recovery Projects. Paper SPE 75148 presented at the SPE/DOE IOR Symposium, Tulsa, Oklahoma, April 13-17, 2002.

Hudson J.W., Jochen J.E., and Jochen V.A. 2000. Practical Technique to Identify Infill Potential in Low-Permeability Gas ReservoirsApplied to the Milk River Formation in Canada. Paper SPE 59779 presented at the SPE/CERI Gas Technology Symposium, 3-5April, Calgary, Alberta, Canada, 2000.

Hudson J.W., Jochen J.E., and Spivey J.P. 2001. Potential Methods to High-Grade Infill Opportunities Applied to the Mesaverde, Morrowand Cotton Valley Formations. Paper SPE 68598 presented at the SPE Hydrocarbon Economics and Evaluation Symposium, 2-3April, Dallas, Texas, 2001.

Ibatullin, R. R., Ibragimov, N. G., Khisamov, R. S. Podymov, E. D., Shutov, A.A. 2002. Application and method based on artificialintelligence for selection of structures and screening of technologies for enhanced oil recovery. Paper SPE 75175 presented at theSPE/DOE IOR Symposium, Tulsa, Oklahoma, April 13-17, 2002.

Joseph, J., Taber, F., David, M. and Seright R.S. 1996. EOR Screening Criteria Revisited. Paper SPE 35385 presented at the SPE/DOEImproved Oil Recovery Symposium, Tulsa, OK, April 21-24, 1996.

Alberta Research Council (ARC, 2006). Prize Version 3.1 Manual, February 2006.IRIS Research (IRIS, 2007). SWORD Analytical Tool Manual (Version 2.1), October 2007.Mackie, S.I. and Welsh, M.B. 2006. An Oil and Gas Decision-Making Taxonomy. Paper SPE 100699 presented at the SPE Asia Pacific Oil

& Gas Conference and Exhibition, Adelaide, Australia, 11-13 September 2006.Manrique, E., Ranson, A. and Alvarado, V. 2003. Perspectives of CO 2 injection in Venezuela. Paper presented at the 24

th AnnualWorkshop and Symposium for the IEA Collaborative Project on Enhanced Oil Recovery, Canada, September 7-10, 2003.

Manrique, E.J. and Pereira, C.A. Identifying Viable EOR Thermal Processes in Canadian Tar Sands. 2007. Paper 2007-176 presented at thePetroleum Societys 8 th Canadian International Petroleum Conference, Calgary, Alberta, Canada, 12-14 June 2007.

Narayanasamy R., Davies D.R., and J.M. 2006. Somerville, Well Location Selection From a Static Model and Multiple Realisations of aGeomodel Using Productivity-Potential Map Technique. Paper SPE 99877 presented at the SPE Europec/EAGE Annual Conferenceand Exhibition, Vienna, Austria, 12-15 June 2006.

Palmgren, C., Renard, G. 1995. Screening Criteria for the Application of Steam Injection and Horizontal Wells. 8 th EuropeanSymposium on Improved Oil Recovery, Vienna, Austria, May 15-17, 1995.


16/16

16 SPE 113269

Pedersen, F.B., Hanssen, T.H., and Aasheim, T.I. 2006. How Far Can a State-of-the-Art NPV Model Take You in Decision Making? PaperSPE 99627 presented at the SPE Europec/EAGE Annual Conference and Exhibition, Vienna, Austria, 12-15 June 2006.

Skinner, D.C. 2001. Introduction to Decision Analysis. Probabilistic Publishing, Gainsville, FL 32653.Thompson M. A. and Goodyear S. G. 2001. Identifying Improved Oil Recovery Potential: A New Systematic Risk Management. Paper

SPE 72103 presented at the 2001 Asia Pacific Improved Oil Recovery Conference, Kuala Lumpur, October 8-9, 2001.Tyler, N. and Finley, R. J. Architectural Controls on the Recovery of Hydrocarbons from Sandstone Reservoirs. SEPM Concepts in

Sedimentology and Paleontology, (1991), V3, pp: 37.Velasquez, D., Rey, O., and Manrique, E. 2006. An overview of carbon dioxide sequestration in depleted oil and gas reservoirs in Florida,

USGS Petroleum Province 50. Paper presented at the 4th LACCEI International Latin American and Caribbean Conference forEngineering and Technology (LACCEI 2006), Mayagez, Puerto Rico, June 21-23, 2006

Virine, L. and Murphy, D. 2007. Analysis of Multicriteria Decision-Making Methodologies for the Petroleum Industry. Paper IPTC presented at the Internat ional Petroleum Technology Conference, Dubai, U.A.E., 4-6 December 2007.

Voneiff G. W. and Cipolla C. 1996. A New Approach to large-Scale Infill Evaluations Applied to the OZONA (Canyon) Gas. Paper SPE35203 presented at the Permian Basin Oil and Gas Recovery Conference, 27-29 March, Midland, Texas, 1996.

Welsh, M.B. and Begg, S.H. 2007. Modeling the Economic Impact of Cognitive Biases on Oil and Gas Decisions. Paper SPE 110765 presented at the 2007 SPE Annual Technical Conference and Exhibition, Anaheim, California, 11-14 November 2007.

Welsh, M.B., Bratvold, R.B., and Begg, S.H. Cognitive Biases in the Petroleum Industry: Impact and Remediation. 2005. Paper SPE 96423 presented at the 2006 SPE Annual Technical Conference and Exhibition, Dallas, Texas, 9-12 October 2005.

Wozinak D.A., Wing J. L., Schrider L.A. 1997.Infill Reserve Growth Resulting From Gas Huff-n-Puff and Infill Drilling - A Case History.Paper SPE 39214 presented at the SPE Eastern Regional Meeting, 22-24 October, Lexington, Kentucky, 1997.

spe113269 effective eor decision strategies with limited data manrique

Documents